JP2017094055A - Gesture recognition method, apparatus and wearable device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a gesture recognition method, apparatus and wearable device.SOLUTION: A gesture recognition apparatus for recognizing a body motion and controlling a function of a device detects a motion artifact due to an intentional movement of a user from an output signal of a biosignal sensor, and generates a control signal to control a function of a target device that corresponds to a reference signal pattern if a signal pattern of the detected motion artifact corresponds to the reference signal pattern.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an operation recognition method and apparatus, and a wearable apparatus.

  The motion recognition technique is a technique for recognizing movements of all or part of the body such as the user's body, hand, and face to control the function of the device. As a motion recognition technique, a method for detecting a change in an electrical signal due to a muscle movement using an electromyography (EMG) sensor and estimating a user's motion based on the detected change in the electrical signal, and an acceleration sensor In addition, there is a method of measuring the inertia due to the movement of the user's body using a motion sensor such as a gyro sensor, and estimating the user's movement based on the measured inertia.

  The present invention has been made in view of the above prior art, and an object of the present invention is to provide a motion recognition method and device for recognizing body movement and controlling the function of the device, and a wearable device. It is in.

  A motion recognition method for recognizing body movement and controlling the function of the device according to an aspect of the present invention to achieve the above object comprises detecting a motion artifact from an output signal of a biological signal sensor; Generating a control signal for controlling a function of a target device corresponding to the reference signal pattern when the signal pattern of the motion artifact corresponds to a reference signal pattern registered in advance;

The step of generating the control signal determines a type of an action performed by a user based on the signal pattern of the motion artifact and at least one reference signal pattern, and the target corresponding to the determined type of action Control signals that control the function of the device may be generated.
The motion artifact may be caused by a physical pressure or motion that a user applies to the biosignal sensor or the periphery of the biosignal sensor.
The motion recognition method may further include estimating a user's health information based on a biological signal detected from the output signal when no motion artifact is detected from the output signal.
The step of generating the control signal may include a step of performing user authentication based on a biological signal detected from the output signal, and a step of generating the control signal when the user authentication is successful.
The step of generating the control signal may include the step of determining whether or not there is a reference signal pattern corresponding to the signal pattern of the motion artifact among the reference signal patterns registered in advance.
The reference signal pattern may be generated by determining a function type according to a user selection and registering a motion artifact signal pattern corresponding to the determined function type.
The biological signal sensor is included in a wearable device, and the step of generating a control signal for controlling a function of the target device is a step of controlling a function of the wearable device or a function of another device connected to the wearable device. Can be included.
The other device may be any one of a mobile terminal device, an Internet of Things (IoT) device, and a smart car.
The step of detecting the motion artifact may include a step of determining whether the motion artifact has occurred based on an average value of the output signal in a predetermined time interval.
The operation recognition method may further include a step of wirelessly transmitting the control signal to a device that performs the function.
The biological signal sensor may measure a biological signal including at least one of a photoelectric volume pulse wave signal and an electrocardiogram signal, and the biological signal may relate to a user's health condition.

  The motion recognition device for recognizing body movement and controlling the function of the device according to one aspect of the present invention made to achieve the above object stores at least one processor and instructions executed by the processor At least one memory, wherein the processor corresponds to an operation of detecting a motion artifact from an output signal of the biological signal sensor according to the instruction, and a signal pattern of the motion artifact corresponds to a pre-registered reference signal pattern In this case, an operation of generating a control signal for controlling the function of the target device corresponding to the reference signal pattern is executed.

The biological signal sensor may be included in a wearable device, and the operation of generating the control signal may generate a control signal that controls a function of the wearable device or a function of another device connected to the wearable device.
The operation of generating the control signal may include an operation of determining whether or not there is a reference signal pattern corresponding to the signal pattern of the motion artifact among the reference signal patterns registered in advance.

  A wearable device according to an aspect of the present invention made to achieve the above object includes a biological signal sensor for measuring a biological signal, and a motion for determining whether or not a motion artifact is indicated in an output signal of the biological signal sensor. An artifact detector; and a motion recognition device that determines whether a signal pattern of the motion artifact corresponds to a pre-registered reference signal pattern when it is determined that a motion artifact is indicated in the output signal; A health information determiner that processes the output signal of the biological signal sensor to determine the health status of the user when it is determined that no motion artifact is indicated in the output signal.

The motion recognition device determines a type of motion performed by a user based on the signal pattern of the motion artifact and at least one reference signal pattern, and controls a function of the target device corresponding to the determined motion type. A control signal may be generated.
The motion recognition device may perform user authentication based on the biometric signal, and generate the control signal when the user authentication is successful.
The wearable device further includes a control signal generator that generates a control signal based on the reference signal, and a transmitter that transmits the control signal to a control target device located at a location away from the wearable device. it can.

According to the present invention, it is possible to perform motion recognition using a biological signal sensor for measuring a biological signal without a separate additional sensor for motion recognition.
Further, since it is not necessary to use a sensor that contacts the skin for motion recognition, the convenience for the user is improved, and intuitive motion recognition based on motion artifacts becomes possible.

It is a figure which shows the structure of the operation | movement recognition apparatus by one Embodiment. It is a figure which shows the structure of the wearable apparatus by one Embodiment. It is a figure for demonstrating in detail the structure of the wearable apparatus by one Embodiment. It is a figure for demonstrating the change of the output signal of the biomedical signal sensor by user operation by one Embodiment. It is a figure for demonstrating the change of the output signal of the biomedical signal sensor by user operation by one Embodiment. It is a figure for demonstrating the change of the output signal of the biomedical signal sensor by user operation by one Embodiment. It is a figure which shows the implementation example of the function control by user operation by one Embodiment. It is a figure which shows the implementation example of the function control by user operation by one Embodiment. It is a figure which shows the implementation example of the function control by user operation by one Embodiment. It is a figure which shows the implementation example of the function control by user operation by one Embodiment. It is a flowchart for demonstrating operation | movement of the operation | movement recognition method by one Embodiment. 5 is a flowchart for explaining the operation of the wearable device according to the embodiment. It is a figure which shows the structure of the action recognition apparatus by other embodiment.

  The terminology used in this embodiment is merely used to describe a specific embodiment, and is not intended to limit the embodiment. An expression used in the singular encompasses the expression of the plural, unless it has a clearly different meaning in the context. In this specification, terms such as “including” or “having” indicate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification. All other features or numbers, steps, operations, components, parts, or combinations thereof must be understood as not excluding in advance.

  Unless defined differently, all terms used herein, including technical or scientific terms, are the same as commonly understood by one of ordinary skill in the art to which this embodiment belongs. Has meaning. Commonly used predefined terms should be construed as having a meaning consistent with the meaning possessed in the context of the related art and are ideal or excessive unless explicitly defined herein. It is not interpreted as a formal meaning.

  Moreover, when it demonstrates with reference to drawings, the same referential mark is provided to the same component irrespective of drawing code | symbol, and the duplication description with respect to it is abbreviate | omitted. In the case where it is determined that the specific description of the related art in the description of the present embodiment unnecessarily obscure the gist of the embodiment, the detailed description thereof will be omitted.

  The following embodiments are applied to recognize the actions performed by the user and control the functions of one device. According to one embodiment, when a user wearing a wearable device performs a pre-registered motion for motion recognition, motion artifacts due to the user's motion are detected from the output signal of the biological signal sensor included in the wearable device, and detected. The user's motion is recognized based on the signal pattern of the generated motion artifact. As a result of the action recognition, the wearable apparatus or another apparatus connected to the wearable apparatus is controlled so that a function defined in advance by the specific action is executed.

  The wearable device has, for example, a watch, a bracelet, a band, glasses, a ring, and the like. Devices connected to the wearable device include mobile terminals, smart home appliances, Internet of Things (IoT) devices, smart cars, and the like, but the type of device is not limited thereto. The mobile terminal is, for example, a personal information terminal (PDA) having a communication function, a smartphone, a tablet computer, a laptop computer, or the like, and a smart home appliance is a TV, a refrigerator, an audio, an air conditioner, a washing machine, a set Top box, vacuum cleaner, etc. The Internet of Things device includes all devices that can be remotely controlled based on the Internet.

  Hereinafter, specific examples of embodiments for carrying out the present invention will be described in detail with reference to the drawings. The same reference numerals shown in the drawings indicate the same members.

  FIG. 1 is a diagram illustrating a configuration of a motion recognition apparatus according to an embodiment.

  The motion recognition device 110 recognizes a user's motion. The motion recognition device 110 recognizes the user's motion based on the output signal of the biological signal sensor 120 that measures the user's biological signal. According to one embodiment, the motion recognition device 110 and the biological signal sensor 120 are included in a wearable device. However, the scope of the embodiment is not limited to this, and as another example, the motion recognition device 110 and the biological signal sensor 120 are included in separate devices. The motion recognition device 110 recognizes the user's motion based on a change in the output signal of the biological signal sensor 120 indicated when the user wearing the wearable device performs a specific motion. Here, the biological signal means a signal including biological information measured from the user's body, such as a photoplethysmogram (PPG) or an electrocardiogram (ECG). The PPG signal indicates a change in blood vessel volume due to a change in blood flow, and the ECG signal indicates a change in cardiac potential. Hereinafter, the operation of the motion recognition device 110 will be described in detail based on the configuration of the motion recognition device 110.

  Referring to FIG. 1, the motion recognition device 110 includes a motion artifact detector 130, a reference signal pattern storage 140, a motion type determiner 150, and a control signal generator 160. The reference signal pattern storage 140 has a memory including a reference signal pattern.

  The user performs a specific operation to execute a predefined function, and a motion artifact is shown in the output signal of the biological signal sensor 120 by the user's operation. The movement artifact means noise generated by the user's movement when the biological signal is measured from the user's body by the biological signal sensor 120. The motion artifact is not a signal generated by movement due to muscle contraction or relaxation, but is generated by physical pressure or movement that the user intentionally applies to or around the biological signal sensor 120. When the user performs a predefined motion for motion recognition, the output signal of the biological signal sensor 120 includes an intentional motion artifact, and the signal pattern of the motion artifact varies depending on the motion performed by the user. It is.

  The motion artifact detector 130 detects a motion artifact from the output signal of the biological signal sensor 120. The motion artifact detector 130 determines whether the output signal of the biological signal sensor 120 indicates a biological signal measured from the user's body or indicates a motion artifact due to user action. According to one embodiment, the motion artifact detector 130 determines that a motion artifact has occurred when a signal having an amplitude greater than or equal to a threshold is indicated from the output signal of the biological signal sensor 120. The threshold is determined based on the average value of the output signal of the biological signal sensor 120 in a predetermined time interval. For example, the threshold value increases as the average value of the output signal increases, and the threshold value decreases as the average value decreases.

  According to another embodiment, the motion artifact detector 130 samples the output signal of the biological signal sensor 120 at a predetermined time period and determines an average value of the sampled signal values. The motion artifact detector 130 obtains a deviation between the determined average value and the signal value of the sampled output signal, and when the magnitude of the deviation is larger than a preset threshold, the output of the biological signal sensor 120 Determine that motion artifacts have occurred in the signal.

  If a motion artifact is detected, the motion type determiner 150 determines the type of motion performed by the user based on the signal pattern of the motion artifact and one or more reference signal patterns registered in advance. The signal pattern of the motion artifact is determined based on, for example, how many times the signal waveform of the motion artifact swings up and down in the time interval in which the motion artifact is indicated, or how many times it swings. When there are a plurality of reference signal patterns, the operation type determiner 150 determines whether there is a reference signal pattern corresponding to the motion artifact signal pattern from the reference signal patterns. The operation corresponding to each reference signal pattern is defined in advance, and the operation type determiner 150 determines the type of the operation by identifying the reference signal pattern corresponding to the signal pattern of the motion artifact.

  The reference signal pattern is registered through a separate registration process, and one or more reference signal patterns are included depending on the number of registered user actions. According to one embodiment, the registration process of the reference signal pattern includes a process of determining a device and a function type to be controlled by a user and registering a motion artifact signal pattern corresponding to the determined device and function type. Including. The user attempts to register and stores the signal pattern of the motion artifact shown in the output signal of the biological signal sensor 120, thereby registering the reference signal pattern corresponding to the user action. Information regarding the registered reference signal pattern is stored in the reference signal pattern storage 140.

  The motion type determiner 150 compares the feature of the signal pattern of the motion artifact with the feature of the reference signal pattern registered in advance, and determines whether the signal pattern of the motion artifact corresponds to the reference signal pattern. The feature of the signal pattern of the motion artifact is extracted based on the waveform form of the motion artifact signal, the feature point (eg, peak point, maximum point, minimum point) of the motion artifact signal. For example, when the similarity between the waveform form of the motion artifact signal and the waveform form indicated by the reference signal pattern is equal to or greater than a threshold, the motion type determiner 150 determines that the motion artifact signal pattern corresponds to the reference signal pattern. decide. As another example, the motion type determiner 150 uses, as a reference signal pattern corresponding to a motion artifact signal pattern, a reference signal pattern having the same number as the number of peak points greater than or equal to the threshold value indicated in the signal waveform of the motion artifact. decide.

  The control signal generator 160 generates a control signal for controlling the function of the target device corresponding to the user action determined by the action type determiner 150. The control signal indicates a signal for controlling a predefined function for one apparatus, and the function controlled by the control signal is determined in advance by user selection during the registration process of the reference signal pattern. The control signal is transmitted to the device to be controlled that performs the function.

  According to another embodiment, the motion recognition device 110 further includes a transmitter 170 that wirelessly transmits the control signal generated by the control signal generator 160 to the control target device. When the control target device is spatially separated from the motion recognition device 110, the transmitter 170 wirelessly transmits a control signal to the control target device. Here, the control signal is configured in a packet form, and the address information of the device to be controlled, information on the function to be controlled, and the like are included in the payload in the packet.

  As described above, the motion recognition apparatus 110 does not use a separate sensor for motion recognition (for example, an electromyogram sensor or a motion sensor), and the biosignal sensor 120 for measuring a biosignal. Motion recognition can be performed based on the change in the output signal. By not using a separate sensor for motion recognition, power consumption is reduced and manufacturing costs are reduced. When an electromyogram sensor is used for motion recognition, the electromyogram sensor must be attached to the skin in order to detect the movement of the muscle. According to this embodiment, a PPG sensor that does not adhere to the skin is used. Since the operation can be recognized, the convenience for the user is improved. In addition, according to the present embodiment, it is possible to recognize the motion more intuitively than in the case of using an electromyogram sensor that detects a minute muscle electrical change.

  In addition, the motion recognition method using an electromyogram sensor or motion sensor recognizes the motion by detecting the movement of the muscle or the movement of the device, so the motion is misrecognized in the process of performing one motion unintentionally by the user. May be. However, according to the present embodiment, motion artifacts generated by the user moving the biological signal sensor with intention or applying physical pressure to the peripheral area of the biological signal sensor are used for motion recognition. The probability of misrecognizing an action different from that decreases.

  FIG. 2 is a diagram illustrating a configuration of a wearable device according to an embodiment.

  The wearable device 200 can monitor the health state of the user and can easily control the functions of the peripheral devices through operation recognition. The wearable device 200 determines the user's health information based on a biological signal (eg, PPG signal, ECG signal) measured using the biological signal sensor 210, or is indicated in the output signal of the biological signal sensor 210. Recognize user movement based on motion artifacts.

  Referring to FIG. 2, the wearable device 200 includes a biological signal sensor 210, a motion artifact detector 220, a health information determiner 230, and a motion recognition device 240.

  The motion artifact detector 220 detects a motion artifact from the output signal of the biological signal sensor 210. For the motion artifact detector 220, the description for the motion artifact detector 130 described in the embodiment shown in FIG. When the motion artifact is not detected from the output signal of the biological signal sensor 210, the motion artifact detector 220 transmits the biological signal that is the output signal of the biological signal sensor 210 to the health information determiner 230, and when the motion artifact is detected. Transmits the output signal of the biological signal sensor 210 to the motion recognition device 240.

  The health information determiner 230 determines the user's health information based on the biological signal measured by the biological signal sensor 210. According to one embodiment, when the biological signal sensor 210 measures a PPG signal, the health information determiner 230 estimates cardiovascular information such as blood pressure based on the PPG signal. According to another embodiment, when the biological signal sensor 210 measures an ECG signal, the health information determiner 230 estimates heart disease information based on the ECG signal.

  The motion recognition device 240 performs motion recognition based on the motion artifact signal pattern detected by the motion artifact detector 220. The motion recognition device 240 estimates the type of user motion by comparing the signal pattern of the motion artifact and the reference signal pattern, and generates a control signal for controlling the function of the target device corresponding to the user motion.

  A detailed description of the health information determiner 230 and the motion recognition device 240 will be described below with reference to FIG. As described above, the wearable device 200 determines the user's health information based on the biological signal measured by the biological signal sensor 210 in a normal state, and when the user performs an intentional operation for function control, the biological signal sensor The signal pattern of the motion artifact shown in the output signal 210 is analyzed to generate a control signal that controls the function of the target device corresponding to the user's motion.

  FIG. 3 is a diagram for explaining the configuration of the wearable device 200 according to an embodiment in more detail.

  Referring to FIG. 3, the health information determiner 230 includes a biological signal processor 350 and a health information estimator 360. The biological signal processor 350 processes the biological signal and extracts signal features from the biological signal. The signal features include, for example, a maximum point, a minimum point, a peak point, an inflection point, a slope maximum point, a slope minimum point, a signal waveform area, and the like of the biological signal waveform, but the type of the signal feature is limited to this. There is nothing. The health information estimator 360 estimates health information related to the user's health based on the signal characteristics of the biological signal. The health information estimated by the health information estimator 360 is provided to the user using an output interface (not shown) such as a display or a speaker, or is stored in a separate storage (not shown). If the health information estimator 360 monitors the user's health based on the estimated health information and determines that an abnormality has occurred in the user's health, the health information estimator 360 provides a warning notification to the user.

  The motion recognition device 240 includes a motion type determiner 320, a reference signal pattern storage 310, and a control signal generator 330. The motion type determiner 320 determines the type of user motion based on the motion artifact signal pattern and one or more reference signal patterns stored in the reference signal pattern storage 310. The control signal generator 330 generates a control signal corresponding to the determined operation. Here, a control signal for controlling the function of the wearable device 200 or the function of another device connected to the wearable device 200 is generated. For each of the operation type determiner 320 and the control signal generator 330, the description of the operation type determiner 150 and the control signal generator 160 described in the embodiment shown in FIG. . According to one embodiment, the motion recognition device 240 further includes a transmitter 340, which transmits the control signal to the controlled device wirelessly.

  According to another embodiment, the motion recognition device 240 further includes a user authenticator 370. The user authenticator 370 performs user authentication to determine whether or not the user who currently wears the wearable device 200 is a registered user based on the biological signal measured using the biological signal sensor 210. The user authenticator 370 determines that the user authentication is successful when the measured signal characteristic of the biological signal corresponds to the registered signal characteristic of the user's biological signal, and the signal characteristic of the registered user's biological signal is determined. If not, it is determined that user authentication has failed. When the user authentication is successful, the user authenticator 370 performs control so that the control signal is normally output from the control signal generator 330, and when the user authentication fails, the control signal is not output from the control signal generator 330. To control.

  4A to 4C are diagrams for explaining changes in the output signal of the biological signal sensor due to a user operation according to an embodiment. 4A to 4C, as one embodiment, a biological signal sensor that measures a PPG signal from a user's wrist is provided in a wristwatch-type wearable device, and the types of user actions registered in advance are the first action, the second action, And 3 of the third action.

  FIG. 4A shows a case where the user grasps the wearable device and shakes left and right as the first operation. Reference numeral 410 indicates an output signal waveform of the biological signal sensor. When the output signal waveform 410 is observed over time, the PPG signal is measured from the biological signal sensor in the normal first time interval 415. When the user performs the first operation in the second time interval 420 for function control, as shown in the figure, the output signal waveform 410 shows a motion artifact in the form of shaking many times with a large amplitude. In the third time interval 425 after the first operation is finished, the PPG signal measured by the biological signal sensor is again shown in the output signal waveform 410.

  FIG. 4B shows a case where the user pushes the wearable device with another hand while applying a downward force as the second operation. When the output signal waveform 430 of the biological signal sensor is observed over time, the PPG signal is measured in the first time interval 435. In the second time interval 440 in which the user performs the second operation, the signal of the output signal of the biological signal sensor A motion artifact is shown in which the value remains small until the end of the second motion. In the third time period 445 after the second operation is finished, the PPG signal is again shown in the output signal waveform 430.

  FIG. 4C shows a case where the user strikes the area adjacent to the wearable device twice with a finger as the third operation. When the output signal waveform 450 of the biological signal sensor is observed over time, the PPG signal is measured from the biological signal sensor in the first time interval 455, and in the second time interval 460 in which the user performs the third operation, as illustrated in FIG. A motion artifact in the form of shaking twice with large amplitude is shown. In the third time interval 465 after the third operation is finished, the output signal waveform 450 indicates the PPG signal again.

  4A to 4C, the wearable device determines whether or not motion artifacts are indicated in the output signal of the biological signal sensor. If it is determined that no motion artifacts are indicated, the wearable device is based on the PPG signal. Estimate health information. When it is determined that a motion artifact is indicated in the output signal of the biological signal sensor, the wearable device determines the type of motion performed by the user based on the motion artifact signal pattern, and the target device corresponding to the user motion A control signal for performing the function is generated. Depending on the motion performed by the user, the signal pattern of the motion artifact is shown differently, and the type of motion performed by the user is determined by distinguishing the signal pattern of the motion artifact according to each motion. .

  The types of user actions described with reference to FIGS. 4A to 4C are not limited by the above-described embodiments, and various modifications and variations are possible.

  FIG. 5A to FIG. 5D are diagrams illustrating an implementation example of function control by a user operation according to an embodiment. In the embodiment shown in FIGS. 5A to 5D, it is assumed that the function control is executed based on the first operation, the second operation, and the third operation described in the embodiment shown in FIGS. 4A to 4C. .

  FIG. 5A shows a case where the control target device is a wearable device. When the first operation 510 executed by the user is recognized, a function of moving the screen of the wearable device to the home screen is executed. If the second operation 520 is recognized, a toggle function to turn off / on the wearable device is performed. When the third operation 530 is recognized, an application frequently used in the wearable device or an application registered in advance by the user is executed.

  FIG. 5B shows a case where the device to be controlled is a smartphone and a music application is currently being executed. When the first operation 510 is recognized, a function for playing the next song is executed by the music application, and when the second operation 520 is recognized, a toggle function for pausing / replaying music playback is executed. If the third action 530 is recognized, a function is executed to increase (or decrease) the audio volume by one step in the music application. Since the user can control the function of the application by intuitive motion recognition instead of directly operating the application executed on the smartphone by touch input, the convenience of the user is improved.

  FIG. 5C shows a case where the device to be controlled is a smart lighting device among the Internet devices of things. If the first action 510 is recognized, a specific function of the smart lighting fixture (eg, change in lighting color) is performed. If the second action 520 is recognized, a function to turn on / off the smart lighting fixture is performed. If the third action 530 is recognized, a function to increase (or decrease) the brightness of the smart lighting fixture is performed. The user can easily control the functions of the peripheral device by recognizing the operation using the wearable device.

  FIG. 5D shows a case where the device to be controlled is a smart car. When the first operation 510 is recognized, the function of opening the door of the smart vehicle is executed in the traveling stop state. When the second operation 520 is recognized, the smart vehicle is started and the function of operating the door lock device is executed. When the third operation 530 is recognized, a function of opening the trunk of the smart vehicle in the traveling stop state is executed. According to one embodiment, user authentication is performed based on the biological signal before function control is performed. For example, when user authentication is performed to determine whether the user wearing the wearable device is a registered user based on the user's biosignal measured by the biosignal sensor and the user authentication is successful, the operation is recognized. The function control by is executed. In this case, motion recognition based on the motion artifact signal pattern is performed prior to user authentication, or user authentication is performed prior to motion recognition.

  In each embodiment of Drawing 5B-Drawing 5D, the control signal generated by motion recognition is transmitted to a smart phone, smart lighting equipment, and a smart car by radio. Here, the control signal is transmitted directly from the wearable device to the control target device, or is transmitted to the control target device via another device (for example, a smartphone) connected to the wearable device. The wearable device and the control target device are, for example, in accordance with a communication system based on WiFi, NFC (Near-Field Communication), 2G (generation) or 3G cellular communication system, LTE (Long Term Evolution), Bluetooth (registered trademark). The control signal is transmitted from the wearable device to the control target device by the communication method.

  The types of operations described with reference to FIGS. 5A to 5D and the function of the target device corresponding to each operation are not limited by the above-described embodiments, and various modifications and variations are possible.

  FIG. 6 is a flowchart for explaining the operation of the motion recognition method according to the embodiment. Referring to FIG. 6, in step S610, the motion recognition apparatus detects a motion artifact from the output signal of the biological signal sensor. In step S620, the motion recognition apparatus determines whether the signal pattern of the motion artifact corresponds to a predefined reference signal pattern. If the signal pattern of the motion artifact corresponds to the reference signal pattern, the motion recognition device generates a control signal for controlling the function of the target device corresponding to the reference signal pattern in step S630. The control signal is transmitted to the control target device that performs the function, and the control target device performs the function defined by the received control signal. When the control target devices are spatially separated, in step S640, the motion recognition device selectively transmits a control signal to the control target device wirelessly.

  The detailed description of the motion recognition method shown in FIG. 6 has been described above with reference to FIG.

  FIG. 7 is a flowchart for explaining the operation of the wearable device according to the embodiment. Referring to FIG. 7, in step S710, the wearable device receives an output signal of the biological signal sensor. In step S720, the wearable device determines whether a motion artifact is detected from the output signal of the biological signal sensor. If a motion artifact is detected, in step S730, the wearable device compares the reference signal pattern registered in advance with the signal pattern of the motion artifact to determine the type of operation performed by the user. In step S740, the wearable device generates a control signal that controls the function of the target device corresponding to the type of operation, and the control signal is transmitted to the control target device that performs the function. If the control target device is spatially separated, in step S750, the wearable device selectively transmits a control signal to the control target device wirelessly.

  If no motion artifact is detected in step S720, the wearable device extracts a biological signal feature from the output signal of the biological signal sensor in step S760. The biological signal feature means a signal feature of a biological signal to be measured by the biological signal sensor. In step S770, the wearable device estimates the user's health information based on the biological signal characteristics.

  Since the detailed description of the motion recognition method shown in FIG. 7 has been described above with reference to FIGS.

  FIG. 8 is a diagram illustrating a configuration of a motion recognition apparatus according to another embodiment.

  Referring to FIG. 8, the motion recognition apparatus 800 includes one or more processors 810 and a memory 820.

  The processor 810 performs one or more operations described above with reference to FIGS. For example, the processor 810 detects a motion artifact from the output signal of the biological signal sensor, estimates the user motion based on the detected motion artifact signal pattern, and then controls a predefined function corresponding to the user motion. An operation for generating a control signal is executed. Such a processor 810 is implemented as a plurality of logic gate arrays, but it should be understood by those having ordinary knowledge in the technical field to which the present embodiment belongs that other forms of hardware can be implemented. Can do.

  The memory 820 stores an instruction for performing one or more operations described above with reference to FIGS. 1 to 7, and stores data and results acquired with the operation of the operation recognition apparatus 800. In one embodiment, the memory 820 is a non-transitory computer readable recording medium, such as high speed random access memory and / or non-volatile computer readable storage medium (eg, one or more disk storage devices, flash memory). Device, or other non-volatile solid-state memory device).

  The motion recognition apparatus 800 according to the present embodiment further includes an input / output interface 830 such as a display, a keyboard, a touch screen, and a microphone, and a network communication interface (not shown) for communicating with the outside. However, the scope of the present embodiment is not limited to this, and as one embodiment, the processor 810 uses the output signal of the biological signal sensor as an input. The input / output interface 830 receives a user input and outputs a result of operation recognition (for example, information on a function to be executed). The network communication interface transmits a control signal generated by the processor 810 to a device external to the motion recognition device 800.

  The above-described embodiments are implemented by hardware components, software components, or a combination of hardware components and software components. For example, the apparatus and components described in the present embodiment include, for example, a processor, a controller, an ALU (arithmetic logic unit), a digital signal processor (digital signal processor), a microcomputer, an FPA (field programmable array), and a PLU (programmable logarithm). It may be implemented using one or more general purpose or special purpose computers, such as units, microprocessors, or different devices that execute and respond to instructions. The processing device executes an operating system (OS) and one or more software applications running on the operating system. The processing device also accesses, stores, manipulates, processes, and generates data in response to software execution. For convenience of understanding, one processing device may be described as being used, but those having ordinary knowledge in the art may recognize that the processing device has multiple processing elements and / or It can be seen that it includes multiple types of processing elements. For example, the processing device includes a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

  The software includes computer programs, code, instructions, or a combination of one or more of these, configures the processing device to operate as desired, and instructs the processing device independently or in combination. Software and / or data is interpreted by the processing device and any type of machine, component, physical device, virtual device, computer storage medium or device for providing instructions or data to the processing device, or transmitted signal It can be realized permanently or temporarily via waves. The software is distributed on computer systems connected to the network and stored or executed in a distributed manner. Software and data are stored on one or more computer-readable recording media.

  The method according to the present embodiment is embodied in the form of program instructions executed through various computer means, and is recorded on a computer-readable recording medium. The recording medium includes program instructions, data files, data structures, etc. alone or in combination. The recording medium and the program instructions may be specially designed and configured for the purpose of the present invention, and may be known and usable by those skilled in the art having computer software technology. . Examples of computer-readable recording media include magnetic media such as hard disks, floppy (registered trademark) disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magnetic-lights such as floppy disks. Media and hardware devices specially configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language code generated by a compiler but also high-level language code executed by a computer using an interpreter or the like. A hardware device may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

  As mentioned above, although embodiment of this invention was described in detail, referring drawings, this invention is not limited to the above-mentioned embodiment, In the range which does not deviate from the technical scope of this invention, it changes variously. It is possible to implement.

110, 240, 800 Motion recognition device 120, 210 Biological signal sensor 130, 220 Motion artifact detector 140, 310 Reference signal pattern storage 150, 320 Motion type determiner 160, 330 Control signal generator 170, 340 Transmitter 200 Wearable device 230 Health information determiner 350 Biological signal processor 360 Health information estimator 370 User authenticator 410, 430, 450 Output signal waveform 415, 435, 455 First time interval 420, 440, 460 Second time interval 425, 445, 465 Third Time interval 510 First operation 520 Second operation 530 Third operation 810 Processor 820 Memory 830 Input / output interface

Claims (20)

A motion recognition method for recognizing body movements and controlling device functions,
Detecting motion artifacts from the output signal of the biological signal sensor;
Generating a control signal for controlling the function of the target device corresponding to the reference signal pattern when the signal pattern of the motion artifact corresponds to a reference signal pattern registered in advance. Method.   The step of generating the control signal determines a type of an action performed by a user based on the signal pattern of the motion artifact and at least one reference signal pattern, and the target corresponding to the determined type of action The operation recognition method according to claim 1, wherein a control signal for controlling a function of the apparatus is generated.   The motion recognition method according to claim 1, wherein the motion artifact is generated by a physical pressure or a motion that a user applies to the biological signal sensor or a periphery of the biological signal sensor.   4. The method according to claim 1, further comprising estimating user health information based on a biological signal detected from the output signal when no motion artifact is detected from the output signal. 5. The operation recognition method described. Generating the control signal comprises:
Performing user authentication based on a biological signal detected from the output signal;
The operation recognition method according to claim 1, further comprising: generating the control signal when the user authentication is successful.   The step of generating the control signal includes a step of determining whether or not there is a reference signal pattern corresponding to the signal pattern of the motion artifact among the reference signal patterns registered in advance. The motion recognition method according to any one of 1 to 5.   7. The reference signal pattern according to claim 1, wherein a function type is determined by user selection, and a signal pattern of a motion artifact corresponding to the determined function type is registered. The motion recognition method according to any one of the above. The biological signal sensor is included in a wearable device,
The step of generating a control signal for controlling a function of the target device includes a step of controlling a function of the wearable device or a function of another device connected to the wearable device. The operation recognition method according to any one of the above.   9. The method according to claim 8, wherein the other device is one of a mobile terminal device, an Internet of Things (IoT) device, and a smart car.   10. The step of detecting the motion artifact includes a step of determining whether or not the motion artifact has occurred based on an average value of the output signal in a predetermined time interval. The operation recognition method according to claim 1.   The operation recognition method according to claim 1, further comprising a step of wirelessly transmitting the control signal to a device that performs the function. The biological signal sensor measures a biological signal including at least one of a photoelectric volume pulse wave signal and an electrocardiogram signal;
The motion recognition method according to claim 1, wherein the biological signal relates to a user's health condition.   A computer-readable recording medium having recorded thereon a program for causing a computer to execute the operation recognition method according to any one of claims 1 to 12 in combination with hardware. A motion recognition device for recognizing body movements and controlling device functions,
At least one processor;
At least one memory storing instructions to be executed by the processor;
The processor is responsive to the instruction.
Detecting motion artifacts from the output signal of the biological signal sensor;
An operation for generating a control signal for controlling a function of a target device corresponding to the reference signal pattern when the signal pattern of the motion artifact corresponds to a reference signal pattern registered in advance; Recognition device. The biological signal sensor is included in a wearable device,
The operation recognition apparatus according to claim 14, wherein the operation of generating the control signal generates a control signal for controlling a function of the wearable apparatus or a function of another apparatus connected to the wearable apparatus.   The operation of generating the control signal includes an operation of determining whether or not there is a reference signal pattern corresponding to the signal pattern of the motion artifact among the reference signal patterns registered in advance. The motion recognition device according to 14 or 15. A biological signal sensor for measuring a biological signal;
A motion artifact detector for determining whether a motion artifact is indicated in the output signal of the biological signal sensor;
A motion recognition device that determines whether or not a signal pattern of the motion artifact corresponds to a pre-registered reference signal pattern when it is determined that a motion artifact is indicated in the output signal;
A health information determinator that processes an output signal of the biological signal sensor to determine a health condition of a user when it is determined that no motion artifact is indicated in the output signal. apparatus.   The motion recognition device determines a type of motion performed by a user based on the signal pattern of the motion artifact and at least one reference signal pattern, and controls a function of the target device corresponding to the determined motion type. The wearable device according to claim 17, wherein the control signal is generated.   The wearable device according to claim 17 or 18, wherein the motion recognition device performs user authentication based on the biometric signal, and generates the control signal when the user authentication is successful. A control signal generator for generating a control signal based on the reference signal;
The wearable device according to any one of claims 17 to 19, further comprising a transmitter that transmits the control signal to a control target device located at a location away from the wearable device.

Images